Learning objectives

• Produce boxplots, scatter plots and smoothed plots using ggplot.

• Describe what faceting is and apply faceting in ggplot.

• Modify the aesthetics of an existing ggplot plot (including axis labels and color).

• Build complex and customized plots from data in a data frame.

Building your plots iteratively

• Building plots with ggplot2 is typically an iterative process.

• We start by defining the dataset we’ll use, lay out the axes, and choose a geom:

• Then, we start modifying this plot to extract more information from it.

• For instance, we can add transparency (alpha) to avoid overplotting:

• We can also add colors for all the points:

• Or to color each species in the plot differently, you could use a vector as an input to the argument color.

• ggplot2 will provide a different color corresponding to different values in the vector. Here is an example where we color with species_id:

• Load required package

library(tidyverse)

• Set the directory

setwd("~")

• Load the data

bw_df <- read.csv("Data/birthweight2.csv")
names(bw_df)

##  [1] "id"      "matage"  "ht"      "gestwks" "sex"    
##  [6] "bweight" "ethnic"  "lbw"     "agegrp"  "lbw2"   
## [11] "agegrp1"

Boxplot - for a categorical and continuous variable

• We will use boxplots to visualize the distribution of birth weight by gender:

• Boxplots provides a standardized way of displaying the distribution of data

• It attempts to provide a visual shape of the data distribution.

• This is based on some summary measures: min, \(1^{st}\) quartile, median, \(3^{rd}\) quartile, max

• Range, IQR, Outliers - 3 \(\times\) IQR above \(3^{rd}\) or below \(1^{st}\) quartiles.

Lets do a Box plot?

• A box plot of bweight vs sex

ggplot(data = bw_df)

Adding aesthetics and labels

ggplot(data = bw_df) + geom_boxplot(aes(y = bweight, x = sex, 
    fill = sex)) + ylab("Birth Weight") + xlab("Sex") + ggtitle("B weight vs Sex")

Box plot and add scatter

• By adding points to the boxplot, we can have a better idea of the number of measurements and of their distribution:

ggplot(data = bw_df, mapping = aes(y = bweight, x = sex, fill = sex)) + 
    geom_boxplot() + geom_point() + ylab("Birth Weight") + xlab("Sex") + 
    ggtitle("Birth weight vs Sex")

Box plot and add scatter points that are jittered

• We will jitter points to reduce overplotting

• Notice how the boxplot layer is behind the jitter layer?

• What do you need to change in the code to put the boxplot in front of the points such that it’s not hidden?

ggplot(data = bw_df, mapping = aes(y = bweight, x = sex, fill = sex)) + 
    geom_boxplot() + geom_point() + geom_jitter() + ylab("Birth Weight") + 
    xlab("Sex") + ggtitle("Birth weight vs Sex")

• Boxplots are useful summaries, but hide the shape of the distribution.

• For example, if there is a bimodal distribution, it would not be observed with a boxplot.

• An alternative to the boxplot is the violin plot (sometimes known as a beanplot), where the shape (of the density of points) is drawn.

• Replace the box plot with a violin plot; see geom_violin().

Violin plot

ggplot(data = bw_df, mapping = aes(y = bweight, x = sex, fill = sex)) + 
    geom_violin() + geom_point() + geom_jitter() + ylab("Birth Weight") + 
    xlab("Sex") + ggtitle("Birth weight vs Sex")

Scatter plot with ggplot2 - for two continuous variables

• We can build a plot sequentially to see how each grammatical layer changes the appearance

• Start with data and aesthetics

# Start with data and aesthetics
ggplot(data = bw_df, mapping = aes(x = gestwks, y = bweight, 
    color = sex))

Add a point geom

# Start with data and aesthetics
ggplot(data = bw_df, mapping = aes(x = gestwks, y = bweight, 
    color = sex)) + # Add a point geom
geom_point()

Add a smooth geom

• To add a regression line on a scatter plot, use the function geom_smooth()

# Start with data and aesthetics
ggplot(data = bw_df, mapping = aes(x = gestwks, y = bweight, 
    color = sex)) + # Add a point geom
geom_point() + ## Add a smooth geom
geom_smooth()

## `geom_smooth()` using method = 'loess' and formula 'y ~ x'

Make the smooth geom straight

• geom_smooth() is used in combination with the argument method = lm. lm stands for linear model.

# Start with data and aesthetics
ggplot(data = bw_df, mapping = aes(x = gestwks, y = bweight, 
    color = sex)) + # Add a point geom
geom_point() + ## Add a smooth geom geom_smooth() + Make it straight
geom_smooth(method = "lm")

## `geom_smooth()` using formula 'y ~ x'

Faceting

• ggplot has a special technique called faceting that allows the user to split one plot into multiple plots based on a factor included in the dataset.

• We will use it to make a scatter plot of birth weight vs gestwks stratified by gender:

• Now we would like to split each plot by the sex of each individual measured.

• You can also organise the panels only by columns (or only by rows):

Facet by sex

# Start with data and aesthetics
ggplot(data = bw_df, mapping = aes(x = gestwks, y = bweight, 
    color = sex)) + # Add a point geom
geom_point() + ## Add a smooth geom geom_smooth() + Make it straight
geom_smooth(method = "lm") + # Facet by sex
facet_wrap(vars(sex), ncol = 2)

## `geom_smooth()` using formula 'y ~ x'

add labels

# Start with data and aesthetics
ggplot(data = bw_df, mapping = aes(x = gestwks, y = bweight, 
    color = sex)) + # Add a point geom
geom_point() + ## Add a smooth geom geom_smooth() + Make it straight
geom_smooth(method = "lm") + # Facet by sex
facet_wrap(vars(sex), ncol = 2) + ## add labels
labs(x = "Gestation weeks", y = "Birthweight", color = "Sex", 
    title = "Lower gestation weeks leads to low birthweight", 
    subtitle = "Birth weight is in grams", caption = "Is there something we can observe by gender")

## `geom_smooth()` using formula 'y ~ x'

adding ggplot2 themes

## `geom_smooth()` using formula 'y ~ x'

Exporting plots

my_plot2 <- ggplot(data = bw_df, mapping = aes(x = gestwks, y = bweight, 
    color = sex)) + geom_point() + geom_smooth(method = "lm") + 
    facet_wrap(vars(sex), ncol = 2) + labs(x = "Gestation weeks", 
    y = "Birthweight", color = "Sex", title = "Lower gestation weeks leads to low birthweight", 
    subtitle = "Birth weight is in grams", caption = "Is there something we can observe by gender") + 
    theme_bw() + theme(plot.title = element_text(size = 15, face = "bold"), 
    axis.text.x = element_text(size = 8), axis.text.y = element_text(size = 8), 
    axis.title.x = element_text(size = 10), axis.title.y = element_text(size = 10)) + 
    scale_color_discrete(name = "Sex")
ggsave("Output/Bweight.pdf", my_plot2, width = 15, height = 10)

## `geom_smooth()` using formula 'y ~ x'

my_plot2

## `geom_smooth()` using formula 'y ~ x'

Useful link and resource with examples and code

https://www.data-to-viz.com/

Break out session - Exercises

• Replicate the plot below and save

ggplot(data = bw_df, mapping = aes(x = gestwks, y = bweight, 
    color = sex)) + geom_point() + geom_smooth(method = "lm") + 
    facet_wrap(vars(sex), ncol = 2) + labs(x = "Gestation weeks", 
    y = "Birthweight", color = "Sex", title = "Lower gestation weeks leads to low birthweight", 
    subtitle = "Birth weight is in grams", caption = "Is there something we can observe by gender") + 
    theme_bw() + theme(plot.title = element_text(size = 15, face = "bold"), 
    axis.text.x = element_text(size = 8), axis.text.y = element_text(size = 8), 
    axis.title.x = element_text(size = 10), axis.title.y = element_text(size = 10)) + 
    scale_color_discrete(name = "Sex")

## `geom_smooth()` using formula 'y ~ x'

• Can you make the shape of the points in the scatter plot to vary with ethnicity?

• add a scale shape attribute. Hint use: scale_shape_discrete(name=“legend title”)

• Instead of having multiple smoothing lines for each ethnic group, integrate them all under one line.

Solution

## `geom_smooth()` using formula 'y ~ x'